The role of IgG4 (+) plasma cells in the association of
Hashimoto’s thyroiditis with papillary carcinoma
FUNDA TAS�LI,1,2 G €ULIZ €OZK€OK,1 ASUMAN ARGON,1 DIDEM ERS€OZ,1 AYS�E YA �GCI,1
ADAM USLU,3 NAZIF ERKAN,3 TARIK SALMAN4 and ENVER VARDAR1
1Department of Pathology, Bozyaka Training and Research Hospital, Izmir; 2Department of Pathology,Faculty of Medicine, Sifa University, Izmir; 3Department of General Surgery, Bozyaka Training and
Research Hospital, Izmir; and 4Department of Medical Oncology, Bozyaka Training and Research Hospital,Izmir, Turkey
Tas�li F, €Ozk€ok G, Argon A, Ers€oz D, Ya�gci A, Uslu A, Erkan N, Salman T, Vardar E. The role of IgG4 (+)plasma cells in the association of Hashimoto’s thyroiditis with papillary carcinoma. APMIS 2014.
Hashimoto’s thyroiditis (HT) is considered to be a risk factor for the formation of papillary carcinoma. The associationof IgG4-related sclerosing disease with tumor is reported to be as sporadic cases in many organs. In this study, it wasintended to re-classify the HT diagnosed cases on the basis of the existence of IgG4 (+) plasma cells; to investigate theclinicopathologic and histopathologic features of the both groups; and in addition, to evaluate the papillary carcinomaprevalence in IgG4 (+) and IgG4 (�) HT cases as well as the prognostic parameters between these groups. Totally 59cases between the years 2008–2013, 29 of which contain Hashimoto thyroiditis diagnosis in total thyroidectomy materi-als, and 30 of which contain the diagnosis of HT+papillary carcinoma, were included in the study. The materials wereimmunohistochemically applied IgG and IgG4; and the cases were classified in two groups as IgG4-positive HT andIgG4-negative HT containing cases, on the basis of IgG4/IgG rate. All histopathologic and clinicopathologic parame-ters between these two groups, as well as their association with papillary carcinoma were investigated. Thirty eight(64.4%) of total 59 cases were NonIgG4 thyroiditis, and 21 (35.5%) were IgG4 thyroiditis. Tumors were detected in 14(36.8%) of the NonIgG4 thyroiditis cases, and in 16 (76.1%) of the IgG4 thyroiditis cases. The association of IgG4thyroiditis with tumor is statistically significant (p < 0.004). Multifocality was found to be at a higher rate in IgG4 thy-roiditis cases. Perithyroidal extension was detected in six of the cases with tumor, and five of the six cases were IgG4thyroiditis cases. The association of IgG4 (+) HT cases with increased papillary carcinoma prevalence is suggestive ofthat IgG4 (+) plasma cells can play a role in carcinogenesis in papillary carcinomas developed in HTs, without achronic sclerosing ground. In addition, although the number of cases is limited, the high-association of IgG4 (+) plasmacells with adverse prognostic parameters such as multifocality and extrathyroidal extension is attention-grabbing. Torender these possibilities evaluable, studies to be carried out with larger case series are needed.
Funda Tas�li, Department of Pathology, Faculty of Medicine, Sifa University, Izmir, Turkey. e-mail: [email protected]
Thyroiditis refers to the existence of inflammationin thyroid gland; and it is a wide group ofdiseases including various inflammatory diseases.Histologically, thyroiditis is sub-classified as acutesuppurative thyroiditis, subacute thyroiditis,Hashimoto’s thyroiditis (HT), focal lymphocyticthyroiditis, palpation thyroiditis, and Riedel’s thy-roiditis. Among these, HT is clinically character-ized by goiter and high level of serum thyroidautoantibody (1). This disease was first defined as
‘struma lymphomatosa’ in 1912 by FokoukaHashimoto, who is a Japan surgeon; and it wasasserted to be a disease different from Grave’sdisease and Riedel’s thyroiditis. Now, HT is con-sidered to be a form of autoimmune thyroiddisease (2). HT is the most common form of auto-immune thyroiditis, and affects 2% of the generalpopulation (2). It is 5–10 times more common inwomen than in men (2).
Although the diagnostic criteria are well-definedin this disease, HT is characterized by a differentclinical course and clinical presentation, and patho-genesis could not be understood well (3).Received 17 March 2014. Accepted 4 June 2014
It is defined as a ‘IgG4-related sclerosing disease(IgG4-RSD)’ that appeared as a multisystemdisease recently. IgG4-RSD is a fibroinflammatorydisease characterized by lymphoplasmacytic infiltra-tion rich in IgG4-positive plasma cells, storiformfibrosis, and – frequently but not always – highserum IgG4 level in different organs (4). This dis-ease has been first identified in autoimmune pancre-atitis by Hamano et al. (5). However, besidespancreas, it has also been identified in some otherorgans such as gallbladder, liver, salivary gland,retroperitoneum, lung, kidney, prostate, orbita, andlymph node (1, 6). In recent years, Li et al. definedIgG4 thyroiditis as a variant of HT (1, 3, 7). IgG4might be a form of generalized IgG4-RSD or thy-roiditis or a form of autoimmune thyroiditis adja-cent to an organ (2).
The association of HT with carcinoma was firstdefined in literature in 1995 by Dailey et al. (8). Inthe literature, the association of IgG4-related scle-rosing disease with tumor is reported to be as spo-radic cases in many organs; and the association ofIgG4-related HT with thyroid papillary carcinomais defined in only one case (9–11).
In this study, the cases to which thyroidectomyhave been applied due to thyroiditis or suspicion ofmalignancy, and which have been diagnosed withHT, will be investigated retrospectively. In thisstudy, it was intended to re-classify the cases on thebasis of the existence of IgG4 (+) plasma cells; toinvestigate the histopathologic features of the bothgroups; and in addition, to evaluate the papillarycarcinoma prevalence in IgG4 (+) and IgG4 (�)HT cases as well as the clinicopathologic and prog-nostic parameters between these groups.
MATERIALS AND METHODS
Totally 59 cases between the years 2008 and 2013 wereincluded in the study, which had been diagnosed with HTin the surgery materials after bilateral total thyroidectomyoperation. The cases were taken from the pathologyrecords of Izmir Bozyaka Training and Research Hospital,and IgG4 concentrations could not be measured becauseof the fact that the serum samples of the cases have notbeen stored. Twenty nine of the 59 cases included in thestudy were HT cases, and the remaining 30 were casesdiagnosed with HT + papillary carcinoma. Eleven of thecases were prediagnosed with malignancy, and the other49 cases were subjected to operation as a result of the ini-tial diagnosis of multinodular goiter. In all cases, in whichhigh serum thyroid autoantibodies were detected, HT washistopathologically diagnosed – whether or not thereexisted formation of lymphoid follicles – when thereexisted lymphoplasmacytic inflammatory infiltration, andoncocytic change in follicular epithelial cells. Tissue sam-ples obtained from the cases were fixed with formalin andwere embedded in paraffin blocks. The sections prepared
from paraffin blocks were applied routine Hematoxylin–Eosin (HE) dye as well as Rabbit polyclonal anti-humanIgG (Dako, Glostrup, Denmark, dilution 1/500) andmouse monoclonal anti-human IgG4 (Abcam, Cambrigde,UK, dilution: 1/500) antibodies were applied with immu-nohistochemical methods. With HE sections, in each caseparameters such as the amount of inflammation, oncocyticepithelial changes were examined histopathologically, andwere scored semi-quantitatively as 0; negative, +1; mild,+2; moderate, +3; severe. In addition, existence andabsence of stromal fibrosis was histopathologically noted.Immunohistochemically applied IgG and IgG4 (+) plasmacells were counted in the areas where the most intense col-oring was available. In both antibodies, the positive cellsin three major magnification areas were counted and aver-aged. Afterward, the cases were classified in two groups asIgG4-positive HT (IgG4 thyroiditis) and IgG4-negativeHT (NonIgG4 thyroiditis), on the basis of IgG4/IgG rate(1, 3, 6). A magnification area in the examination is0.057 mm2 (Olympus BX43 microscope, 910 eyepiece,940 lens; Olympus DP73 camera and Olympus CellsensEntry software, Tokyo, Japan). All histopathologic andclinicopathologic parameters between these two groupswere investigated, and the association of the groups withthyroid was investigated, as well.
Data are shown as the arithmetic mean � SD. Statisti-cal analyses were performed with the Mann–WhitneyU-test and v2 test or Fisher’s exact probability test. Analy-ses were performed and p < 0.05 was considered as statis-tically significant.
RESULTS
In totally 59 HT cases included in the study, fourof the patients were male and 55 were female; andthe median age was 49.6 (min: 28; max: 80). Thecases were sub-classified as IgG4 thyroiditis andNonIgG4 thyroiditis cases, by the acception ofCheuk et al.’s IgG4/IgG rate greater than 40%. Byusing these criteria, 21 IgG4 thyroiditis and 38 No-nIgG4 thyroiditis cases were defined in the study.Intense IgG4 (+) plasma cell infiltration and higherIgG4/IgG rate were identified in IgG4 thyroiditiscases (Figs 1–4), when compared to the NonIgG4thyroiditis cases (Table 1). Statistical significancewas detected between the two groups, in terms ofIgG4 (+) plasma cell number (p < 0.000). In con-trast, statistical difference was not found in IgG4thyroiditis and NonIgG4 thyroiditis cases, in termsof IgG (+) plasma cell number (p = 0.362). Stromalfibrosis was found to be 90.4% (19 of 21 cases) and65.7% (25 of 38 cases) in IgG4 thyroiditis andNonIgG4 thyroiditis cases, respectively. There is astatistical significance between the two groups interms of stromal fibrosis development (p = 0.034).However, any statistical significance was not foundin histopathologic parameters such as lymphoplas-macytic infiltration and oncocytic changes betweenthese two groups.
In the study, there are totally 30 cases having theassociation of Papillary Thyroid Carcinoma withHT (Figs 5 and 6). Clinicopathologic parameters ofthe cases with tumor are summarized in Table 2.Tumor was detected in 16 of 21 cases (76%) and 14of 38 cases (37%) in IgG4 thyroiditis cases andNonIgG4 cases, respectively. The association ofIgG4 thyroiditis with tumor is statistically signifi-cant (chi-square) (p < 0.004). The median ages intumor and IgG4 thyroiditis and tumor andNonIgG4 thyroiditis cases were 45 � 12.1 and46 � 13.9, respectively. The average tumor diame-ter in IgG4 thyroiditis and HT cases was 1.35 cm,and 44% (7 of 16 cases) were papillary microcarci-noma; while the average tumor diameter in No-nIgG4 thyroiditis and HT cases was 1.17 cm, and
58% (8 of 14 cases) were papillary microcarcinomacases. Statistical significance was not detectedbetween the two groups, in terms of tumor size.IgG4 (+) plasma cells in the cases with the associa-tion of HT with tumor were monitored both in in-tratumoral and peritumoral localizations. IgG4 (+)plasma cells were with peritumoral localization in10 of 16 cases (62.5%) in IgG4 thyroiditis, whilethey were with peritumoral localization in 10 of 14cases (71.4%) in NonIgG4 thyroiditis. Both follicu-lar and classic papillary carcinomas were detectedin IgG4 thyroiditis and NonIgG4 thyroiditis cases.The tumors in 69% (11 of 16 cases) of the IgG4thyroiditis cases; and 50% of the NonIgG4 thyroid-itis cases (7 of 14 cases) were in classic type of pap-illary carcinoma morphology. Multifocality was
detected in 9 of 30 cases with tumor. Six (67%) ofthe cases, in which multifocality was detected, arethe cases with IgG4 thyroiditis; and three (33%)are the cases with NonIgG4 thyroiditis; and mul-tifocality was found to be at a higher rate in IgG4thyroiditis cases (Fisher’s exact test) (p = 0.29).Perithyroidal extension was detected in six of thecases with tumor, and five of the six cases wereIgG4 thyroiditis cases (p = 0.209). The cases wereclassified in accordance with the AMES criteria.Twenty eight of the 30 tumor cases were in low-riskgroup, while one patient with high risk was in thy-roiditis and one was in NonIgG4 thyroiditisgroups.
DISCUSSION
Hashimoto’s thyroiditis is the most common formof autoimmune thyroid diseases, and affects 2% ofthe general population (2, 7). It is more frequentlyseen in advanced age women, and 10 times morecommon in women than in men (2, 6). Diagnostichistopathologic features of HT are diffuse lympho-plasmacytic infiltration and formation of lymphoidfollicles with evident germinal centers, as well ashurtle cell changes characterized by large eosino-philic granular cytoplasm in thyrocytes (7). Fibrosisat variable rates is seen in HT; however, this is nota diagnostic criterion (7).
IgG4-RSD that was defined in recent years andinvolves many organs, is a fibroinflammatory dis-ease characterized by lymphoplasmacytic infiltra-tion rich in IgG4-positive plasma cells, storiformfibrosis, and – frequently but not always – highserum IgG4 level in different organs (4). IgG4-RSDhas been first defined by Hamano et al. as autoim-mune pancreatitis (5), and then entity has been
Table 2. Clinicopathologic parameters in patients withtumor (n = 30)
Fig. 6. Tumor tissue which have fibrous capsule andpapillary structures, separated from the surrounding thy-roid tissue. The lymphoid follicles which have prominentgerminal centers in the surrounding thyroid tissue (9100,HE).
identified in many organs (7). In the literature, it isasserted that serum IgG4 levels might be normal inIgG4-RSD cases diagnosed by biopsy (12). For thefirst time, HT was sub-classified by Li et al. in2009, on the basis of IgG4 (+) plasma cells, asIgG4 thyroiditis and NonIgG4 thyroiditis (1). Lateron, the same research staff made an analysis con-taining also clinical findings and serum IgG4 levelsin IgG4 thyroiditis and NonIgG4 thyroiditis sub-groups in larger case series. In consequence of theanalysis, they concluded that these two groups werecompletely different entities and that IgG4 thyroid-itis might not be a component of generalized IgG4-RSD (3). In a study, Kakudo et al. said that IgG4thyroiditis might have two subtypes. The authorsspeculated these two subtypes in such a way thatthyroid involvement may appear as Reidel’s thy-roiditis if IgG4-RSD is in the form of systemic dis-ease; and thyroid involvement may appear as HTtype if IgG4-RSD is specific to the organ. Serumsamples of the cases could not be reached in ourretrospective study that we presented. As the SerumIgG4 levels of the events were unknown, any com-ment could not be made about whether or notIgG4 thyroiditis cases were the component of sys-temic IgG4-RSD; and with the immunohistochemi-cal data, the cases were evaluated as a diseaseadjacent to the organ.
In this study, the cases were classified as IgG4thyroiditis and NonIgG4 thyroiditis, with the crite-ria offered by Li et al. first time in their study in2009 (1). In the both cases, stromal fibrosis wasevaluated and found to be very intense in the caseswith IgG4 thyroiditis, compared to the other group;and it was also found to be statistically significant.In the literature, it is asserted that TGFb may playa key role in the relation between IgG4-positivecells and fibrosis because of the fact that TGFb is astrong fibrogenesis stimulus and also has the poten-tial of enhancing the IgG4 formation in B cells (9,13). Also in our HT cases, fibrosis seen in all IgG4-related diseases in all organs was found to be statis-tically significant, as consistent with the literature.
In our study, histologic parameters such as lym-phoid follicle formation and oncocytic change seenin other HT diagnostic criteria, as well as clinicalparameters such as the course and symptoms of thedisease were evaluated; and any statistical signifi-cance could not be found between IgG4 thyroiditisand NonIgG4 thyroiditis groups. However, in theliterature, these two entities are evaluated as differ-ent entities, in terms of clinic, demographic, andsonographic data (2, 3, 7).
Normally, HT is treated medically; however, ifthere are clinical symptoms such as swallowing dif-ficulty, pain, and dyspnea or if there is a suspected
tumor, surgery is indicated (1). HT is mostlydetected in thyroidectomies made upon tumor indi-cation; and in literature, the association of papillarythyroid carcinoma with HT is reported to be at afrequency of 11–36% (2). In the literature, thisassociation was first defined in 1955 (8).
Studies carried out in recent years show that thisassociation cannot be explained as the coexistenceof two frequent entities; and that there might be acause and effect relationship between these twoentities (2). In the literature, the association ofIgG4-RSD with tumor is reported in a limitednumber of documents (9–11, 14). However, therelationship between cancer-related immunity andIgG4 antibody reaction has yet to be fully under-stood (15). In recent years, studies on the effect ofIgG4 (+) cells on cancer tissue were published (16,17). IgG4 antibodies are thought to increase intumor-inducing Th2-associated inflammatory condi-tions; and these antibodies are thought to cause theloss of anti-tumor immunity, by inhibiting theimmune effector cell activation. In their study car-ried out with 54 extrahepatic cholangiocarcinomacases, Harada et al. detected high level of IgG4 (+)and IL-10-related regulatory cytokine, and theyasserted that cancer cells directly cause an increasein IgG4 antibodies by the agency of IL-10-relatedcytokine, like cells providing nonprofessional anti-gen; and that IgG4 reaction may enable tumor toavoid immune surveillance, if the primary role ofIL-10 is considered to be repressing the immuneresponse (17). In their study, Zen et al. showed thatin IgG4-related autoimmune pancreatocholangitiscases, immune pathogenesis was with Th2-domi-nated reaction mediator (18).
In IgG4-positive antibodies and tumors immun-pathogenesis, T regulatory (Treg) cells are the othernoticeable cells besides Th2 cells. As one of thesub-groups controlling the immune system, Tregcells contribute to immunosuppression, by increas-ing the regulatory cytokine such as IL-10 or TGFb(19). Tregs also increase the transformation of Bcells into IgG4-secreting plasma cells (20). In theirstudy investigating the existence of IgG4-positiveplasma cells, Kimura et al. found a positive rela-tionship between the existence of IgG4 and the exis-tence of Treg cells. Like the other authors, theyasserted that Treg cells cause progressive prolifera-tion and differentiation of IgG4-positive cells; andbased on this, IgG4 reaction in cholangiocarcinomamay mediate the suppression of tumor-related pro-gression and tumor progression by the agency ofTreg cells (15). Hiroaki et al. reported that Tregcells have a role in the formation of pancreatic can-cer, and that such cells are associated with poorprognosis (21).
Publications, by which also the relationshipbetween thyroid carcinoma and Treg cells besidespancreatic cancer are investigated, are available (19,22, 23). In the study, in which Gogali et al. investi-gated papillary carcinoma and nodular goiter cases,they ascertained that the cases with tumor had highTreg positivity, and identified positive correlationwith advanced disease (23). In this study, theauthors did not identify any difference in terms ofthe existence of Treg between HT and PapillaryThyroid Carcinoma cases and Papillary ThyroidCarcinoma cases (23). However, in some studies,the existence of Treg is reported to be more fre-quent in Papillary Thyroid Carcinoma cases on thebasis of HT. French et al. associated Tregs withadvanced stage tumors, while Cunha et al. associ-ated Treg positivity with low tumor aggressivenessand good clinical course (19, 22). And Modi et al.concluded that in the papillary carcinoma casesthey investigated in young adult and children agegroups the immune reaction to tumor is bothimportant and complex reaction, and immune sys-tem shows its effect by being active in several direc-tions (24).
In our series, a higher level of tumor associationwas detected in the IgG4 thyroiditis cases, whencompared to NonIgG4 thyroiditis cases, and therewas a statistical significance (p < 0.05). To the bestof our knowledge, the study presented is the firststudy in the literature, in which the association ofIgG4 antibodies with thyroid papillary carcinomawas investigated. The low number of cases andunknown serum IgG4 levels of the cases were arestrictive cause in this study. If the findings can besupported with studies to be carried out with largercase series, the serum IgG4 follow-ups in cases,HTs of which are sub-classified on the basis ofIgG4, can be used as an additional method besidesthe monitoring methods intended for closer follow-up of the association of HT with Papillary ThyroidCarcinoma, based on the findings of Li et al. (3).
In the study presented, the tumors in the IgG4thyroiditis and NonIgG4 thyroiditis cases wereevaluated in terms of various prognostic parameterssuch as tumor size, extrathyroidal extension, mul-tifocality, and tumor stage. When the averagetumor size in IgG4 thyroiditis was compared to theaverage tumor size in the other group, tumors inIgG4 thyroiditis cases tended to be in a diametergreater than 1 cm, while tumors in NonIgG4 groupappeared mostly as microcarcinoma. In otherwords, tumors in IgG4 thyroiditis tend to be in lar-ger diameters. Extrathyroidal extension and mul-tifocality were at a higher rate in tumors in IgG4thyroiditis cases; and there was a statistical signifi-cance between this and the other group. In
addition, the tumor cases in the study were gradedaccording to the AMES criteria. However, in thestudy population, only two tumor cases were in thehigh risk group, while one high risk tumor was inIgG4 thyroiditis group, and the other high risktumor was in NonIgG4 thyroiditis group.
With these findings, we can make the commentthat the tumors in IgG4 thyroiditis cases includedin the study were accompanied with poor prognos-tic parameters. In the literature, in studies carriedout with large case series, the HT and PapillaryThyroid Carcinoma association is reported to beaccompanied by a good prognosis, as discordantwith our findings (22, 24–26). Recently, studies arebeing carried out, which report that IgG4-positivecells and existence of tumor-related lymphocyte,with or without HT as a basis, are accompanied bypoor prognostic parameters, as similar to our find-ings (19). Also in our study, IgG4-positive cells inIgG4 thyroiditis cases with tumor were detectedin both intratumoral and peritumoral areas; and inthese cases, a higher level of perithyroidal exten-sion, multifocality, and larger tumor diameter wereobserved.
In summary, the association of IgG4-RSD withtumor is reported in a limited number of docu-ments in the literature; and the relation betweencancer-related immunity and IgG4 reaction has yetto be understood. In recent years, studies on theeffect of IgG4 (+) cells on cancer tissue were pub-lished. To the best of our knowledge, this study isthe first study in the literature, in which the associa-tion of IgG4 antibodies with thyroid papillary car-cinoma was investigated. In our study, the highincidence of cancer in IgG4 thyroiditis cases wassuggestive of the possible role of IgG4 (+) plasmacells in immune pathogenesis, when compared tothe NonIgG4 thyroiditis cases. In addition,although the number of cases is limited, the high-association of IgG4 (+) plasma cells with adverseprognostic parameters such as multifocality andextrathyroidal extension is attention-grabbing. Torender these possibilities evaluable, prospectivestudies to be carried out with larger case series areneeded.
CONFLICT OF INTEREST
We declare that we have no conflict of interest.
REFERENCES
1. Li Y, Bai Y, Liu Z, Ozaki T, Taniguchi E, Mori I,et al. Immunohistochemistry of IgG4 can help
subclassify Hashimoto’s autoimmune thyroiditis.Pathol Int 2009;59:636–41.
2. Ahmed R, Al-Shaikh S, Akhtar M. Hashimoto thyroid-itis: a century later. Adv Anat Pathol 2012;19:181–6.
3. Li Y, Nishihara E, Hirokawa M, Taniguchi E, Miyau-chi A, Kakudo K. Distinct clinical, serological, andsonographic characteristics of hashimoto’s thyroiditisbased with and without IgG4-positive plasma cells.J Clin Endocrinol Metab 2010;95:1309–17.
4. Deshpande V, Zen Y, Chan JK, Yi EE, Sato Y,Yoshino T, et al. Consensus statement on the pathol-ogy of IgG4-related disease. Mod Pathol 2012;25:1181–92.
5. Hamano H, Kawa S, Horiuchi A, Unno H, FuruyaN, Akamatsu T, et al. High serum IgG4 concentra-tions in patients with sclerosing pancreatitis. N Engl JMed 2001;344:732–8.
6. Kakudo K, Li Y, Hirokawa M, Ozaki T. Diagnosisof Hashimoto’s thyroiditis and IgG4-related sclerosingdisease. Pathol Int 2011;61:175–83.
7. Li Y, Nishihara E, Kakudo K. Hashimoto’s thyroid-itis: old concepts and new insights. Curr Opin Rheu-matol 2011;23:102–7.
8. Dailey ME, Lindsay S, Skahen R. Relation of thyroidneoplasms to Hashimoto disease of the thyroid gland.AMA Arch Surg 1955;70:291–7.
9. Ito M, Naruke Y, Mihara Y, So K, Miyashita T,Origuchi T, et al. Thyroid papillary carcinoma withsolid sclerosing change in IgG4-related sclerosing dis-ease. Pathol Int 2011;61:589–92.
10. Tian W, Yakirevich E, Matoso A, Gnepp DR. IgG4(+) plasma cells in sclerosing variant of mucoepider-moid carcinoma. Am J Surg Pathol 2012;36:973–9.
11. Straub BK, Esposito I, Gotthardt D, Radeleff B,Antolovic D, Flechtenmacher C, et al. IgG4-associ-ated cholangitis with cholangiocarcinoma. VirchowsArch 2011;458:761–5.
12. Sah RP, Chari ST. Serologic issues in IgG4-relatedsystemic disease and autoimmune pancreatitis. CurrOpin Rheumatol 2011;23:108–13.
13. Satoguina JS, Adjobimey T, Arndts K, Hoch J, Old-enburg J, Layland LE, et al. Tr1 and naturally occur-ring regulatory T cells induce IgG4 in B cells throughGITR/GITR-L interaction, IL-10 and TGF-beta. EurJ Immunol 2008;38:3101–13.
14. Witkiewicz AK, Kennedy EP, Kennyon L, Yeo CJ,Hruban RH. Synchronous autoimmune pancreatitisand infiltrating pancreatic ductal adenocarcinoma:case report and review of the literature. Hum Pathol2008;39:1548–51.
15. Kimura Y, Harada K, Nakanuma Y. Pathologicsignificance of immunoglobulin G4-positive plasmacells in extrahepatic cholangiocarcinoma. Hum Pathol2012;43:2149–56.
16. Karagiannis P, Gilbert AE, Nestle FO, KaragiannisSN. IgG4 antibodies and cancer-associated inflamma-tion: insights into a novel mechanism of immuneescape. Oncoimmunology 2013;2:e24889.
17. Harada K, Shimoda S, Kimura Y, Sato Y, Ikeda H,Igarashi S, et al. Significance of immunoglobulin G4(IgG4)-positive cells in extrahepatic cholangiocarcino-ma: molecular mechanism of IgG4 reaction in cancertissue. Hepatology 2012;56:157–64.
18. Zen Y, Fujii T, Harada K, Kawano M, Yamada K,Takahira M, et al. Th2 and regulatory immune reac-tions are increased in immunoglobin G4-related scle-rosing pancreatitis and cholangitis. Hepatology2007;45:1538–46.
19. French JD, Weber ZJ, Fretwell DL, Said S, KlopperJP, Haugen BR. Tumor-associated lymphocytes andincreased FoxP3 + regulatory T cell frequency corre-late with more aggressive papillary thyroid cancer.J Clin Endocrinol Metab 2010;95:2325–33.
20. Koyabu M, Uchida K, Miyoshi H, Sakaguchi Y,Fukui T, Ikeda H, et al. Analysis of regulatory Tcells and IgG4-positive plasma cells amongpatients of IgG4-related sclerosing cholangitis andautoimmune liver diseases. J Gastroenterol 2010;45:732–41.
21. Hiraoka N, Onozato K, Kosuge T, Hirohashi S. Prev-alence of FOXP3 + regulatory T cells increases duringthe progression of pancreatic ductal adenocarcinomaand its premalignant lesions. Clin Cancer Res2006;12:5423–34.
22. Cunha LL, Morari EC, Guihen AC, Razolli D, Ger-hard R, Nonogaki S, et al. Infiltration of a mixture ofimmune cells may be related to good prognosis inpatients with differentiated thyroid carcinoma. ClinEndocrinol (Oxf) 2012;77:918–25.
23. Gogali F, Paterakis G, Rassidakis GZ, Kaltsas G,Liakou CI, Gousis P, et al. Phenotypical analysisof lymphocytes with suppressive and regulatoryproperties (Tregs) and NK cells in the papillary carci-noma of thyroid. J Clin Endocrinol Metab2012;97:1474–82.
24. Modi J, Patel A, Terrell R, Tuttle RM, Francis GL.Papillary thyroid carcinomas from young adults andchildren contain a mixture of lymphocytes. J ClinEndocrinol Metab 2003;88:4418–25.
25. Jeong JS, Kim HK, Lee CR, Park S, Park JH, KangSW, et al. Coexistence of chronic lymphocytic thy-roiditis with papillary thyroid carcinoma: clinical man-ifestation and prognostic outcome. J Korean Med Sci2012;27:883–9.
26. Kim SS, Lee BJ, Lee JC, Kim SJ, Jeon YK, KimMR, et al. Coexistence of Hashimoto’s thyroiditiswith papillary thyroid carcinoma: the influence oflymph node metastasis. Head Neck 2011;33:1272–7.
